Vimal (વિમલ) @101V

imorphr.com Joined April 2009

Tweets

727
Followers

143
Following

1K
Likes

365

Vimal (વિમલ) @101V

a day ago

@FutureStacked A couple of years ago, I was given a brand new surface laptop, I set it up, ran for a few hours and guess what Blue Scree of Death!!! I hope it improves since then

1 0 1 517 0

View Details

#TIL a new word called "Phygital". It is a hybrid word for Digital + Human assistance. In certain country or market segment Digital only is not sufficient you need a human assistant to close the loop. To me the current #AI seems very much "Phygital" where human in the loop is necessity.

0 0 0 8 0

View Details

Vimal (વિમલ) @101V

4 days ago

@goddek @EmmanuelMacron @grok what happened in Paris?

1 0 1 552 0

View Details

Vimal (વિમલ) @101V

a week ago

@itswpceo @grok is this true?

1 0 0 23 0

View Details

Danila Poyarkov @dan_note

a week ago

I extracted Vibe’s TUI into a standalone Elixir package with a perfectly matching name: Cringe 😅 It’s early alpha, but already has layouts, boxes, widgets, Ghostty-backed terminal input, painter diffing, and testable rendering. github.com/elixir-vibe/cr…

0 2 39 2K 15

View Details

Y Combinator @ycombinator

2 weeks ago

In a recent batch talk, YC General Partner @t_blom broke down how to build a self-improving, AI-native company. He walks through how to create recursive, self-improving AI loops, and why founders who get this right will run companies that improve while they sleep. 00:00 — Companies Are Roman Legions 00:54 — Copilots Are the Wrong Mental Model 01:55 — Extract the Domain Knowledge 02:24 — The Recursive Self-Improving Loop 04:12 — The Holy Shit Moment at YC 05:50 — Self-Optimizing Product and Support Loops 06:29 — Burn Tokens, Not Headcount 07:23 — Middle Management Is Over 08:05 — Make Everything Legible to AI 09:40 — Regenerating the YC User Manual 11:19 — Software Is Ephemeral, Context Is Valuable 12:18 — Where Humans Still Matter

76 238 2K 585K 5K

View Details

Vimal (વિમલ) @101V

2 weeks ago

Valid point on the value of dev-to-dev debates, but this overlooks how LLMs shift the learning model. With an LLM, you shift from "learning by debate" to "learning by rapid experimentation." It lets you make mistakes, fail fast, and recover in minutes instead of days.Sometimes the best way to understand why Approach X is bad isn't a 2-hour argument—it's watching the LLM build it in 30 seconds, seeing it break immediately, and learning from the firsthand experience.

0 0 0 135 0

View Details

Vimal (વિમલ) @101V

2 weeks ago

Spending the last 6 months "vibe coding" has completely ruined traditional programming for me. Looking back at writing every line by hand, hunting through docs, and digging in StackOverflow feels as ancient as using punch cards. AI gave the power back to the builder. What a time to be alive. 🚀 Anyone else feel like they can never go back?

0 0 1 14 0

View Details

Vimal (વિમલ) @101V

3 weeks ago

@bcardarella @bcardarella I am genuinely curious about motivation behind zap, are you trying to combine the beautiful syntax from elixir with static type and speed from zig?

1 0 0 72 0

View Details

Vimal (વિમલ) @101V

4 weeks ago

@nikunj @yacineMTB Great work @nikunj, love them all. I am curious about the libs you used for SVG. What have you used for that?

0 0 0 106 0

View Details

Vimal (વિમલ) @101V

a month ago

@GergelyOrosz I wanted to publish an app, paid to Google Playstore, they can't even provide invoice, their support doesn't even know how to do it. The doc is out of date. I have been asked to contact Google Developer Team, they said they are short of staff so cant help...

0 0 0 303 0

View Details

Vimal (વિમલ) @101V

2 months ago

Everything that you hear/read is twisted always take it with a pinch of salt

0 0 0 15 0

View Details

Vimal (વિમલ) @101V

2 months ago

@bcardarella @elixirlang It is with the right feedback loop. In my prompt I always ask it to use feed back from Credo and Dialyzer and the code quality is much better

0 0 0 130 0

View Details

Vimal (વિમલ) @101V

2 months ago

@GeoffreyHuntley Wait until you go to India. UPI is everywhere and Uber works only in major cities and you cannot install the local taxi apps without Indian number

0 0 0 55 0

View Details

gabby @GabriellaG439

3 months ago

New blog post: "A sufficiently detailed spec is code" I wrote this because I was tired of people claiming that the future of agentic coding is thoughtful specification work. As I show in the post, the reality devolves into slop pseudocode haskellforall.com/2026/03/a-suff…

123 263 2K 430K 3K

View Details

Vimal (વિમલ) @101V

3 months ago

A coding question about #elixirlang `String.slice("Elixir", 0..-2//1)` returns "Elixi" but `Enum.map((0..-2//1), &(&1))` returns [], why?

0 0 0 20 0

View Details

Vimal (વિમલ) @101V

3 months ago

When I asked it to look at the code base and make necessary changes to define "Check" field as "boolean" it did partial job and did not take care of the schemas where a default value was specified to be 0 or 1 🤦‍♂️.

0 0 0 8 0

View Details

Vimal (વિમલ) @101V

3 months ago

AI vs. Reality: Claude mapped a smallint checkbox field to a partial index using true/false instead of 0/1. Result? Migration failed. Claude’s investigation? 45 minutes of wrong answers. My investigation? 10 minutes to find and fix. AI is a powerful co-pilot, but the "Human in the Loop" isn't a suggestion—it’s a requirement. 🛠️ #SoftwareEngineering #AI #claude

1 0 0 15 0

View Details

Vimal (વિમલ) @101V

3 months ago

Small types, big headaches. 🧵 I recently hit a snag where Claude AI generated a migration for a PostgreSQL partial index. It treated a smallint checkbox field like a boolean, using 'true'/'false' syntax. The breakdown: The Error: Type mismatch. Postgres doesn't appreciate booleans where integers live. The AI Loop: Claude spent 45+ mins investigating and hallucinating solutions. The Human Fix: 10 mins of manual debugging to spot the type mismatch. LLMs are great at syntax, but they still struggle with the underlying architecture context. We aren't replaceable yet—we're the ones holding the map.